Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Linear systolic arrays for matrix computations

Journal Article · · Journal of Parallel and Distributed Computing; (USA)
;  [1]
  1. Technical Univ. of Munich, Lehrstuhl fur Netzwerktheorie und Schaltungstechnik, Arcisstrasse 21, D-8000 Munich (DE)

The paper presents a systematic method to derive linear processor arrays from given two-dimensional arrays. The proposed hierarchical approach lead to one-dimensional computational arrays for an important class of algorithms which need O(n/sup 3/)computational steps in a sequential implementation. The proposed linear arrays share the following properties: matching of the computation rate and available I/O bandwidth, lexicographical order of input/output data, modularity O(n)computational cells, O(n/sup 2/)delay elements, fault tolerance, and the use of pipelined arithmetic units. As an example, a linear array for matrix inversion is given.

OSTI ID:
5242849
Journal Information:
Journal of Parallel and Distributed Computing; (USA), Journal Name: Journal of Parallel and Distributed Computing; (USA) Vol. 7:1; ISSN JPDCE; ISSN 0743-7315
Country of Publication:
United States
Language:
English

Similar Records

Designing linear systolic arrays
Journal Article · Thu Nov 30 23:00:00 EST 1989 · Journal of Parallel and Distributed Computing; (USA) · OSTI ID:6960988

On mapping algorithms to linear and fault-tolerant systolic arrays
Journal Article · Tue Feb 28 23:00:00 EST 1989 · IEEE Trans. Comput.; (United States) · OSTI ID:6243175

Interstitial fault tolerance-a technique for making systolic arrays fault tolerant
Conference · Fri Dec 31 23:00:00 EST 1982 · OSTI ID:5257924